PCR is a rapid and simple method for specifically amplifying a target DNA sequence in an exponential manner. Saiki, et al. Science 239:487-4391 (1988). Briefly, the method as now commonly practiced utilizes a pair of primers that have nucleotide sequences complementary to the DNA which flanks the target sequence. The primers are mixed with a solution containing the target DNA (the template), a DNA polymerase and dNTPS for all four deoxynucleotides (adenosine (A), tyrosine (T), cytosine (C) and guanine(G)). The mix is then heated to a temperature sufficient to separate the two complementary strands of DNA. The mix is next cooled to a temperature sufficient to allow the primers to specifically anneal to sequences flanking the gene or sequence of interest. The temperature of the reaction mixture is then set to the optimum for the thermophilic DNA polymerase to allow DNA synthesis (extension) to proceed. The temperature regimen is then repeated to constitute each amplification cycle. Thus, PCR consists of multiple cycles of DNA melting, annealing and extension. Twenty replication cycles can yield up to a million-fold amplification of the target DNA sequence. In some applications a single primer sequence functions to prime at both ends of the target, but this only works efficiently if the primer is not too long in length. In some applications several pairs of primers are employed in a process commonly known as multiplex PCR.
The ability to amplify a target DNA molecule by PCR has applications in various areas of technology e.g., environmental and food microbiology (Wernars et al., Appl. Env. Microbiol., 57:1914-1919 (1991); Hill and Keasler, Int. J. Food Microbiol., 12:67-75 (1991)), clinical microbiology (Wages et al. J. Med. Virol., 33:58-63 (1991); Sacramento et al., Mol. Cell Probes, 5:229-240 (1991)), oncology (Kumar and Barbacid, Oncogene, 3:647-651 (1988); McCormick, Cancer Cells, 1:56-61 (1989)), genetic disease prognosis (Handyside et al., Nature, 344:768-770 (1990)), and blood banking and forensics (Jackson, Transfusion, 30:51-57 (1990)).
DNA polymerase obtained from the hot springs bacterium Thermus aquaticus (Taq DNA polymerase) has been instrumental in DNA amplification, DNA sequencing, and in related DNA primer extension techniques. The DNA and amino acid sequences described by Lawyer et al., J. Biol. Chem., 264:6427 (1989), GenBank Accession No. J04639, define the gene encoding Thermus aquaticus DNA polymerase and the enzyme Thermus aquaticus DNA polymerase as those terms are used herein. The highly similar DNA polymerase (Tfl DNA polymerase) expressed by the closely related bacterium Thermus flavus is defined by the DNA and amino acid sequences described by Akhmetzjanov, A. A., and Vakhitov, V. A., Nucleic Acids Research 20: 5839 (1992), GenBank Accession No. X66105. These enzymes are representative of a family of DNA polymerases, also including Thermus thermophilus DNA polymerase, which are thermostable. These enzymes lack a 3'-exonuclease activity such as that which is effective for editing purposes in mesophilic DNA polymerases such as E. coli DNA polymerase I, and phages T7, T3, and T4 DNA polymerases. Thermostable DNA polymerases which exhibit editing function are generally found in thermophilic archaebacteria such as Pyrococcus furiosus. Related DNA polymerases of this class are commonly known as Pfu, Pwo, Pfx, Vent, or Deep Vent.
The availability of thermostable DNA polymerases such as Taq DNA polymerase has both simplified and improved PCR. Taq DNA polymerase is stable up to 95.degree. C. and its use in PCR has eliminated the necessity of repetitive addition of temperature sensitive polymerases after each thermal cycle. Additionally, Taq DNA polymerase can extend DNA at higher temperatures which tends to prevent the non-specific annealing of primers and thus, has improved the specificity and sensitivity of PCR.
Although significant progress has been made in PCR technology, the amplification of non-target oligonucleotides due to side-reactions, such as mispriming on non-target background DNA, RNA, and/or the primers themselves, still presents a significant problem. This is especially true in diagnostic applications where PCR is carried out in a milieu containing complex background DNA while the target DNA may be present in a single copy (Chou et al., Nucleic Acid Res., 20:1717-1723 (1992)).
The temperature at which Taq DNA polymerase exhibits highest activity is in the range 62-72.degree. C.; however, significant activity is exhibited at room temperature, approximately 25.degree. C. to 37.degree. C. In a normal or "cold start," the primers may prime DNA extension at non-specific sequences because the formation of only a few base pairs at the 3'-end of a primer can result in a stable priming complex. The result can be competitive or inhibitory products at the expense of the desired product. As an example of inhibitory product, structures consisting only of primer, sometimes called "primer dimers" are formed by the action of DNA polymerase on primers paired with each other, regardless of the true target template. The probability of undesirable primer-primer interactions increases with the number of primer pairs in the reaction, as with multiplex PCR. During PCR cycling, these non-specific extension products can compete with the desired target DNA.
Further, it has been determined that side reactions often occur when all reactants are mixed at ambient temperature before thermal cycling is initiated. One method for minimizing these side reactions is termed "hot start" PCR. Many PCR analyses, particularly the most demanding ones, benefit from a hot start. About 50% of all PCR reactions show improved yield and/or specificity if a hot start is employed, and in some cases a hot start is absolutely critical. These demanding PCR analyses include those which have very low copy numbers of target (such as 1 HIV genome per 10,000 cells), denatured DNA (many DNA extraction procedures include a boiling step, so that the template is single-stranded during reaction setup), or contaminated DNA e.g., DNA from soil or feces and/or DNA containing large amounts of RNA. However, current methods of achieving a hot start are tedious, expensive, and/or have other shortcomings.
Hot start PCR may be accomplished by various physical, chemical, or biochemical methods. In a physical hot start, the DNA polymerase or one or more reaction components that are essential for DNA polymerase activity is not allowed to contact the sample DNA until all the components required for the reaction are at a high temperature. The temperature must be high enough so that not even partial hybridization of the primers can occur at any locations other than the desired template location, in spite of the entire genome of the cell being available for non-specific partial hybridization of the primers. Thus, the temperature must be high enough so that base pairing of the primers cannot occur at template (or contaminating template) locations with less than perfect or near-perfect homology. This safe starting temperature is typically in the range of 50.degree. C. to 75.degree. C. and typically is about 10.degree. C. hotter than the annealing temperature used in the PCR.
One physical way a hot start can be achieved is by using a wax barrier, such as the method disclosed in U.S. Pat. No. 5,599,660. See also Hebert et al., Mol. Cell Probes, 7:249-252 (1993); Horton et al., Biotechniques, 16:42-43 (1994). Using such methods, the PCR reaction is set up in two layers separated by a 1 mm thick layer of paraffin wax which melts at about 56.degree. C. There are several methods which may be used to separate the reaction components into two solutions. For instance, all of the DNA is added, with 1.times.buffer but no dNTPs and no DNA polymerase enzyme, in a volume of 25 ml. One drop of melted wax is added and the tubes are all heated to 60.degree. C. for one minute to allow the melted wax to form a sealing layer after which the tubes are cooled so the wax solidifies. Then a 25 ml mixture containing 1.times.buffer, all of the dNTPs, and the enzyme is added to each reaction. Finally, 1 drop of oil is added, to make 4 total layers. As the thermal cycler protocol heats the tubes to the first melting step (approximately 95.degree. C.), the wax melts and floats to mix with the oil layer, and the two aqueous layers mix by convection as the temperature cycles.
One common variation involving the use of a wax barrier is that the reaction components are assembled with no magnesium ions so that the DNA polymerase enzyme is inactive. The magnesium ion encased in a wax bead is then (or initially) added. A problem with these wax methods, however, is that the wax hardens after each PCR cycle. This makes sample recovery extremely tedious, since the wax tends to plug the pipet tips used to remove the sample. This is true even if the samples are reheated to melt the wax. Another potential problem is cross-contamination if tweezers are used to add wax beads, since slight contact between the tweezers and the tube caps can move DNA template between samples before the PCR reactions start.
Another way to implement a hot start PCR is to use DNA polymerase which is inactivated chemically but reversibly, such as AMPLITAQ GOLD.TM. DNA polymerase. This enzyme preparation, distributed by PE Applied Biosystems, is distributed to users in inactivated form, but is reactivatable by heating. The required reactivation conditions, however, are extremely harsh to the template DNA: ten minutes at 95.degree. C. and at a nominal pH of 8.3 or lower results in reactivation of some 30% of the enzyme which is enough to start the PCR. See Moretti, et al., BioTechniques 20 25: 716-722 (1998). Because this treatment depurinates DNA every thousand bases or so, this enzyme can not be used to amplify DNA more than a few kilobases in length. Accordingly, the use of this enzyme is most efficient when it is restricted to amplifying target DNA with a length of approximately 200 base pairs.
An additional way of implementing a hot start is to combine the Taq DNA polymerase enzyme with a Taq antibody before adding it to the reagent. This method employs a monoclonal, inactivating antibody raised against Taq DNA polymerase. See Scalice et al., J. Immun. Methods, 172: 147-163 (1994); Sharkey et al., Bio/Technology, 12:506-509 (1994); Kellogg et al., Biotechniques, 16: 1134-1137 (1994). The antibody inhibits the polymerase activity at ambient temperature but is inactivated by heat denaturation once the reaction is thermocycled, thus rendering the polymerase active. Unfortunately, the antibodies currently available for use in this method are not very efficient, and a 5 to 10-fold molar excess must be used to effect the advantages of a hot start PCR. For Klentaq-278, an amino-terminally deleted Thermus aquaticus DNA polymerase that starts with codon 279 which must be used at higher protein levels for long PCR (up to ten times more protein than Taq DNA polymerase), the levels of antibody necessary for a hot start become extremely high and the denatured antibody protein retains some inhibition for longer PCR targets. The original developer of anti-Taq antibodies (Kodak, now Johnson & Johnson) uses a triple-monoclonal antibody mixture which is more effective but is not commercially available and has not been tested in long PCR.
These methods used for hot starts require inclusion of an often expensive component (e.g., anti-Taq antibody) in the reaction mix and may place some undesirable constraints on the performance of the PCR such as a relatively short time period between when a reagent is prepared and when it must be used, or a lower efficiency of amplification. Therefore, it is usually preferable to perform physical hot starts in PCR if at all feasible.
A low tech, inexpensive option is to add the enzyme, the magnesium and/or the dNTPs to the reactions after they have heated up. Besides being tedious and prone to error, this method commonly results in contamination and cross-contamination of PCR samples as the reaction tubes must be opened in the thermal cycler while they are hot.
Some workers believe they are doing a hot start when they set up PCR reactions in tubes on ice, then add the tubes to a thermal cycler block pre-warmed to 95.degree. C. Although some benefit arises from this method, the addition of only a few nucleotides to a primer can take place every second during the fifteen seconds or more that the tubes warm from 0.degree. C. to 25.degree. C. This is enough to initiate unwanted competitive PCR for reactions that require a hot start. Also, if many tubes are involved in an experiment, the tubes placed in the block first are heated for a longer time period at 95.degree. C. compared to the tubes placed later in the heating block thus resulting in a lack of reproducibility between samples.
Thermophilic DNA polymerases are commonly believed to have minimized their mesotemperature activity during their evolution to optimize activity at around 70.degree. C. According to this belief, it should not be possible to further decrease their room temperature activity without seriously compromising either their high temperature activity or their resistance to 95.degree. C.
However, applicants conjectured the possibility that thermostable DNA polymerases could be mutated to a "cold-sensitive" phenotype in order to decrease polymerase activity at room temperature while not harming the activity at the normal optimum extension temperature for PCR, nor the thermostability required for the melting step of each PCR cycle. Such mutants are capable of catalyzing the PCR amplification and exhibiting substantially reduced activity at room temperature, yet near normal activity at optimum reaction temperatures when compared to DNA polymerases without the mutations. Such mutant DNA polymerases would be highly useful in providing a hot-start capability and could be prepared, distributed and used without any additional steps or protocol changes. Thus, by adopting a cold sensitive DNA polymerase, end users could have the advantages of a hot start for all of their PCR analyses, not just the analyses that are first demonstrated to be problematic with a normal room temperature start. Furthermore, "long and accurate" PCR (i.e., employing longer target lengths and with enhanced fidelity) could conveniently be provided the advantages of a hot start without tedious extra care or steps, and human STR typing and multiplex PCR will gain in reliability and efficiency. Such long and accurate PCR is described in Barnes, Proc. Natl. Acad. Sci. USA, 91:2216-2220 (1994) and in U.S. Pat. No. 5,436,149.